A novel hematological score (HS) and its related nomogram model to predict nontuberculous mycobacterial pulmonary disease in patients with suspected multidrug-resistant pulmonary tuberculosis

Autor: Qi Yu, Shengling Hu, Fenfang Liu, Fengyun Gong
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: Annals of Medicine, Vol 56, Iss 1 (2024)
Druh dokumentu: article
ISSN: 07853890
1365-2060
0785-3890
DOI: 10.1080/07853890.2024.2409965
Popis: Background Nontuberculous mycobacteria pulmonary disease (NTM-PD) exhibits clinical and radiological characteristics similar to those of pulmonary tuberculosis (PTB). This study aimed to develop a novel hematological score (HS) and its related nomogram model to identify NTM-PD in patients with suspected multidrug-resistant pulmonary tuberculosis (SMDR-PTB) due to lack of response to first-line anti-TB treatment (ATT).Methods We retrospectively recruited patients with SMDR-PTB from Wuhan Jinyintan Hospital between January 2014 and January 2022. These patients were divided into NTM-PD and MDR-PTB groups based on pathogen test results. Participants were randomly allocated to training and validation set in a 7:3 ratio. The ROC and LASSO regression were employed to select variables. Multivariate logistic analysis was conducted on the training set to develop the HS and its related nomogram models, followed by internal validation on the validation set.Results The HS was constructed and developed on CKMB, ADA, GGT, LDL, and UHR, demonstrating good predictive value with AUCs of 0.900 and 0.867 in the training and validation sets, respectively. The HS-based nomogram model consists of Age, Gender, DM, HIV infection, ILD and HS, and exhibited strong discriminative ability, accuracy, and clinical utility in two sets. The AUCs were 0.930 and 0.948 in the training set and validation set, respectively.Conclusion HS may be a useful biomarker for identifying NTM-PD in patients with SMDR-PTB. The HS-based nomogram model serves as a convenient and efficient tool for guiding the treatment of SMDR-PTB patients.
Databáze: Directory of Open Access Journals